Biology in 4D
July 23rd, 2008 by Yggdrasil
(120 Views)
Hello everyone. Mitch has asked me to contribute to this blog. This may be somewhat difficult as I am a biophysicist which leaves the topic of this blog as the one branch of science left out of the name of my field. Perhaps it would be better if I refer to myself as a biophysical chemist (or would that be physical biochemist? chemical biophysicist?)
Anyway, as the token biologist, I wanted to bring your attention to a commentary in Cell (doi:10.1016/j.cell.2008.06.013) describing the role of Pixar-style computer animations in the future of biology education. Although the article is an interesting read, what I really wanted to show you all is the author’s website (www.molecularmovies.org) which houses a collection of these animations. (Warning: visiting this site can be hazardous to your research productivity)
There are tons of other really great computer animations on the site (though some are not so great in terms of explaining things. Alas, the one video related to organic synthesis falls into the not-so-great category). My personal favorite is the movie on apoptosis (programmed cell death), which features one of my favorite protein complexes, the apoptosome (or as I like to call it, the seven-membered ring of death).
Now, as a biophysicist, I think that these videos are great because they illustrate some very important concepts in biology. The apoptosis video shows how many processes in biology resemble overly complex Rube-Goldberg Machines. Other videos on the site, especially those by Drew Barry, offer a glimpse into an important field of research: protein dynamics. Chemists are used to thinking of catalysts as fairly static entities. Sometimes a catalyst can be as simple as a surface that acts binds a reactant and primes it for subsequent reaction. In contrast, the catalysts in biology, enzymes and ribozymes, are rarely static. The video on DNA replication (video available at the WEHI website) shows the dynamic nature of these biological catalysts. The animation shows the E. coli replisome, a large multienzyme complex, as it copies DNA. The components of this complex have a number of different enzymatic activities that all need to be synchronized and coordinated in order for replication to proceed. Despite all these complicated interactions, the E. coli replisome proceeds at a rate of about 1000 nucleotides per second and with an error rate of about 1 per 109 nucleotides.
Of course, one has to remember that these videos are animations, not realistic simulations. While they are based on empirical results (e.g. crystal structures, biochemical assays, single molecule experiments), the animators do take some creative liberties with the videos. For example, I doubt anyone has observed buzzing and clicking sounds that accompany Brownian motion and enzyme catalysis in many of these videos. Indeed, the animators (with good reason) don’t show two concepts that are becoming increasingly important in understanding biological dynamics: the stochasticity of events in the cell (e.g. polymerases don’t move along at a constant rate) and the very crowded environment of the cell.
Most significant, however, is that while many of these videos depict the dynamics of various enzymes, not much is known about the actual motions of these enzymes and enzyme complexes. When biologists discuss conformational changes, these protein movements are often identified by looking at static “snap-shots” of an enzyme in two different stages of a reaction. Rarely are the kinetics of the transition measured directly, and the techniques that can directly observe conformational changes (e.g. Förster resonance energy transfer) give limited spatial information. Furthermore, the single molecule experiments that give arguably the best kinetic information about enzyme catalysis and protein motion often have limited temporal resolution (it’s hard to go below the millisecond time scale). Computational methods (e.g. molecular dynamics) can give detailed videos of molecular motion with both high spatial and temporal resolution, but modern computers can simulate only tens of nanoseconds, orders of magnitude below the timescale of most large protein motions. NMR spectroscopy has the advantage of being able to access a large range of time scales, but NMR measurements are limited to small systems and can access only dynamics of an enzyme in equilibrium. Being able to somehow synthesize and connect the information from timescales ranging from bond rotations and vibrations to conformational change and allostery is a tough task, but doing so may offer huge insights into the fundamental chemical and physical principles governing enzyme catalysis. Recent attempts to do so (Henzler-Wildman et al. 2007, doi:10.1038/nature06407, doi: 10.1038/nature06410) have been very promising, though there is still much work left for us biophysicists.
It also had a dramatic speed increase over the original designs, at a whopping 18s half-life at the fastest step. Even if it wasn’t going to be turning any relevant loads any time soon, it was a dramatic improvement over the original concept 4 years earlier. Still, even though some clever O-chem tricks made the motor better, it still operated on the same 4-step cycle that Koumura’s did back in 99′. Even recently, in Org. Biomol. Chem., 2008, 6, 507 - 512, DOI: 10.1039/b715652a, Pollard et al. report on substituting naphthalene moieties for phenyl moieties, in order to better control the speed of the motors, and to enable the design and synthesis of more complex systems.
on